Header logo is ei

Planning in Decentralized POMDPs with Predictive Policy Representations

2008

Conference Paper

ei


We discuss the problem of policy representation in stochastic and partially observable systems, and address the case where the policy is a hidden parameter of the planning problem. We propose an adaptation of the Predictive State Representations (PSRs) to this problem by introducing tests (sequences of actions and observations) on policies. The new model, called the Predictive Policy Representations (PPRs), is potentially more compact than the other representations, such as decision trees or Finite-State Controllers (FSCs). In this paper, we show how PPRs can be used to improve the performances of a point-based algorithm for DEC-POMDP.

Author(s): Boularias, A. and Chaib-Draa, B.
Book Title: MASPLAN 2008
Journal: Proceedings of ICAPS 2008 Multiagent Planning Workshop (MASPLAN 2008)
Pages: 1-7
Year: 2008
Month: September
Day: 0

Department(s): Empirical Inference
Bibtex Type: Conference Paper (inproceedings)

Event Name: ICAPS 2008 Multiagent Planning Workshop
Event Place: Sydney, Australia

Digital: 0
Language: en
Organization: Max-Planck-Gesellschaft
School: Biologische Kybernetik

Links: PDF

BibTex

@inproceedings{6836,
  title = {Planning in Decentralized POMDPs with Predictive Policy Representations},
  author = {Boularias, A. and Chaib-Draa, B.},
  journal = {Proceedings of ICAPS 2008 Multiagent Planning Workshop (MASPLAN 2008)},
  booktitle = {MASPLAN 2008},
  pages = {1-7},
  organization = {Max-Planck-Gesellschaft},
  school = {Biologische Kybernetik},
  month = sep,
  year = {2008},
  month_numeric = {9}
}